Designer Daniel Voshart used machine learning to bring ancient statues to life, transforming the stone busts of 54 long-dead Roman emperors into photorealistic faces.
Voshart used a combination of software and other sources to create the portraits. The main tool was ArtBreeder, an online program which uses a machine learning method known as a generative adversarial network (GAN) to manipulate portraits and landscapes.
Voshart fed ArtBreeder images of emperors he collected from statues, coins, and paintings, and then tweaked the portraits manually based on historical descriptions, feeding them back to the GAN.
Voshart originally made 300 posters of the portraits, hoping they'd sell in a year. Instead, they were gone in three weeks.
From The Verge
View Full Article
No entries found